Instructions to use deepseek-ai/DeepSeek-V4.1-Flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deepseek-ai/DeepSeek-V4.1-Flash with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="deepseek-ai/DeepSeek-V4.1-Flash")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-V4.1-Flash", device_map="auto")

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use deepseek-ai/DeepSeek-V4.1-Flash with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deepseek-ai/DeepSeek-V4.1-Flash"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/DeepSeek-V4.1-Flash",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/deepseek-ai/DeepSeek-V4.1-Flash

SGLang

How to use deepseek-ai/DeepSeek-V4.1-Flash with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "deepseek-ai/DeepSeek-V4.1-Flash" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/DeepSeek-V4.1-Flash",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "deepseek-ai/DeepSeek-V4.1-Flash" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/DeepSeek-V4.1-Flash",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use deepseek-ai/DeepSeek-V4.1-Flash with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/DeepSeek-V4.1-Flash
```

Has the model's alignment with human ethics been strengthened compared to the previous generation?

#17

by likewendy - opened Sep 10

Discussion

likewendy

Sep 10

Has the model's alignment with human ethics been strengthened compared to the previous generation? Are there any prompts that successfully bypass its safeguards?
It supports vision capabilities and offers decent performance; since the model isn't overly large, it would clearly be even better if jailbroken.

AgileTurnip

Sep 10

can you not heretic the model?

likewendy

Sep 11

can you not heretic the model?

I don't understand what you're saying.

jebbam

Sep 11

https://github.com/p-e-w/heretic

AgileTurnip

Sep 11

can you not heretic the model?

I don't understand what you're saying.

https://github.com/p-e-w/heretic

jebbam

Sep 11

•

edited Sep 11

can you not heretic the model?

I don't understand what you're saying.

https://github.com/p-e-w/heretic

It looks like heretic may not work on the Deepseek v4 models:
https://github.com/p-e-w/heretic/issues/310

"Why Manifold-Constrained Hyper-Connections Break Abliteration: A Structural Analysis of DeepSeek-V4"
https://huggingface.co/blog/RadicalNotionAI/mhc-ablation-challenges

That same issue points to abliterated models:
https://huggingface.co/apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8

But that looks like non-heretic code used to abliterate it. And the github links in that model README are now 404, so it is unclear how it was done.

There is this code, but again, doesn't look like it is heretic, or at least not "default" heretic:
https://github.com/drowzeys/DeepSeek-V4-Flash-DSpark-Abliterated-Uncensored-1M-57toks

This one has heretic in the name, but it's not quite clear how it was done:
https://huggingface.co/squanchyzx/DeepSeek-V4-Flash-0731-HERETIC-Abliterated-FP8

Happy hacking...

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment